Incident Metering

[xpatUSA]

In incident light meter measures the amount of light falling on a subject, so if your subject is going to impact the reading.

Pardon me, Manfred, but Bruce might not understand that sentence?

[Manfred M]

Pardon me, Manfred, but Bruce might not understand that sentence?

That is something that you and Bill have already gone over in some detail, so I did not think it was worth repeating.

[xpatUSA]

That is something that you and Bill have already gone over in some detail, so I did not think it was worth repeating.

Oh.

I didn't understand your sentence in full. I got that "In incident light meter measures ..." probably meant "In incident light-metering, the meter measures ..." - but - " so if your subject is going to impact the reading." lost me completely. Perhaps you were typing in a hurry and missed something out?

[DanK]

Dan's advice about metering for the brightest parts of a scene is good enough to start with - but with the caveat that those parts may not necessarily be desirable - for example, harsh bright reflections from bright metal parts (cars, motor-cycles) or from glass (white clouds in window-panes). You could just let those blow ...

Indeed. Sorry about the sloppy wording. What I meant was to take a meter reading of the brightest areas you don't want to clip.

[William W]

. . . Dan, the whole exercise was to see how my incident meter stacked up against the light meter in my camera. . .

If "stacked up against" means "do these provide the same accurate reading" then the test is not a the best way of going about a test. To make a test for that exercise, you need a Photographic Grey Card as the scene to measure.

If "stacked up against" means "investigate the different readings these two tools will give in this particular lighting scenario" then you have made that test.

WW

[pnodrog]

I'm looking forward to the day when the dynamic range of a sensor is so great that I can go back to my good old guessing...

[Manfred M]

Oh.

I didn't understand your sentence in full. I got that "In incident light meter measures ..." probably meant "In incident light-metering, the meter measures ..." - but - " so if your subject is going to impact the reading." lost me completely. Perhaps you were typing in a hurry and missed something out?

Sorry Ted, sloppy writing and proof reading on my part.

What I was attempting to say was that an incident light meter reads the quantity of light that falls on your subject. It will give exactly the same reading regardless of the colour or reflectivity of the subject. This of course is not how the reflective light meter in the camera works.

With a reflective light meter, it measures the amount of light that is reflected by the subject. Fill the camera's view screen with an 18% gray card and one would expect the incident and reflective readings to be about the same.

The reason I suggest the readings should be about the same is that we don't know exactly hoe the camera's metering algorithms work. I have read that some camera manufactures use something other than 18% gray as their measurement standard.

[dje]

Fill the camera's view screen with an 18% gray card and one would expect the incident and reflective readings to be about the same.

The reason I suggest the readings should be about the same is that we don't know exactly hoe the camera's metering algorithms work. I have read that some camera manufactures use something other than 18% gray as their measurement standard.

Indeed. Sometimes people extolling the virtues of incident metering fail to indicate that the incident reading still needs to be translated into the reflected scene that the camera sees. As far as I'm aware, that invloves an assumption that the scene has an average reflectance of 18% and is uniform in all directions of a hemisphere. This must lead to errors under some circumstances.

Dave

[xpatUSA]

<> With a reflective light meter, it measures the amount of light that is reflected by the subject. Fill the camera's view screen with an 18% gray card and one would expect the incident and reflective readings to be about the same.

The reason I suggest the readings should be about the same is that we don't know exactly how the camera's metering algorithms work. I have read that some camera manufacturers use something other than 18% gray as their measurement standard.

Quite so. Many cameras now use Recommended Exposure Index (REI) as opposed to the well-defined saturation-based or the Standard Output Sensitivity. REI allows manufacturers to pick an "ISO" number out of thin air based on "testing"to their satisfaction. Which means that a camera could indeed recommend different settings to an incident meter.

[Digital]

After reading more about spot metering, I am wondering,other than being able to isolate the background, is the advantage over matrix metering.

Bruce

[DanK]

Spot metering allows you more control. You decide what areas to meter, and you decide how far off neutral you want them to be. Matrix metering offers none—the camera decides which areas to meter and how much to weight them. Sometimes that extra control is very useful, sometimes not. A simple example is metering in snowy conditions. Snow shouldn’t be neutral; it should be a stop or two brighter than that. Easy to do with a spot meter. Matrix metering is a bit like AWB—the engineers have gotten it to the point where it’s often right, or close. But it’s going to be wrong sometimes, and then you have to take control, one way or another


Sent from my iPhone using Tapatalk

[Digital]

Thank you Dan for the explanation.

Bruce

[pschlute]

After reading more about spot metering, I am wondering,other than being able to isolate the background, is the advantage over matrix metering.

Bruce

Bruce

An important thing to note about spot metering is that it is not a magic bullet. In a high contrast scene like you encountered, spot metering off the highlights will indeed stop them blowing out, but will also mean that the shadow areas will be very dark. You may be able to bring out detail in the latter in PP, especially if you use raw capture.

Very high contrast scenes demonstrate more a problem for the camera's sensor than the meter. If the dynamic range of the scene is more than the sensor can record, whichever method of metering you use is going to produce a compromised image. Either the highlights will be blown or the shadows will be lacking detail.

While it is generally better to control the highlights and bring out the shadows in post, there are better methods to deal with such scenes.

Most cameras today have HDR modes. This works by taking 3 or 5 images at different exposures and the camera merges them into one image. Or you can bracket the exposures yourself and merge the images in software on your computer (my preferred method). So you could spot meter for the highlights for the first bracket. Then spot meter the shadows for the second. Finally either take an average of the first two or use matrix metering for the third bracket. You will after merging the three have an image that has detail throughout.

Another method is to use a Graduated Neutral Density filter in front of the lens. These work well when there is a clear split between the light and dark areas..... say a sea view where the sky is very bright. You then spot meter for the dark area and the light area. If the difference is 3 stops you grab a 3-stop ND Grad from your bag and use that. Set the exposure in Manual on the camera for the reading you got for the dark area and away you go.

[DanK]

Sorry if my post left an impression inconsistent with what Peter wrote. No form of metering avoids problems of excessive dynamic range. What spot metering offers is just the ability for you to decide which setting to use with a scene that has varying brightness, rather than leaving the task to an algorithm that is applied blindly by the camera. The example of snow is a good illustration. Assume that you have a scene that is mostly snow and sky and that does not exceed the dynamic range of the sensor. Matrix metering, averaging metering, central-weighted averaging metering (different brands offer different options) will all get this wrong, setting the exposure too dark because it will try to make some weighted combination of snow and sky neutral, whereas the snow should be one or two stops brighter than that.

I have had some shots where the lighting exceeded the dynamic range of the sensor--for example, Ted's case of a shot with blown out specular highlights--where spot metering is also useful, even in the absence of bracketing. Then you decided what areas have detail you need to preserve, meter off them, and adjust.

The issue is just control.

[xpatUSA]

After reading more about spot metering, I am wondering what, other than being able to isolate the background, is the advantage over matrix metering.

In addition to the good advice offered so far and after looking up Metering Range with reference to scene brights and darks, I found something against spot metering.

Metering range is specified at 0 to 20 EV in Matrix or Center-Weighted, and 2 to 20 EV in Spot metering mode.

https://www.imaging-resource.com/PRO...00/D7000A5.HTM

Meaning that Spot is not so hot when pointed at shadows. I never knew that. Grump.

Then I looked the metering range for my Sigma SD15 DSLR (separate metering sensor under the mirror). It is 1 to 20 EV but ominously specified at 100 ISO, 50mm, f/1.4 - almost certainly meaning less at other settings but no examples given.

So, more trawling revealed this chart for a Nikon F3 film camera:

https://www.mir.com.my/rb/photograph...l/EV_Chart.pdf

Scary stuff. For example it appears that if you load 400 ASA film and set f/11 - you run out of shadow metering at 9EV which is like twilight.

Another chart with explanations on page 41 here:

http://www.alstonhand.com/ueditor/js...1349373607.pdf

Just when I thought I knew everything about metering ...

[pschlute]

Sorry if my post left an impression inconsistent with what Peter wrote.

It did not to me. You made the point very clearly, which I endorse, that many scenes we point our cameras at, need the photographer to take control and use a different metering method to get the exposure "right" for a specific part (or whole...snow) of the scene.

[xpatUSA]

... Spot is not so hot when pointed at shadows. I never knew that. Grump.

Then I looked the metering range for my Sigma SD15 DSLR (separate metering sensor under the mirror). It is 1 to 20 EV but ominously specified at 100 ISO, 50mm, f/1.4 - almost certainly meaning less at other settings but no examples given.

So, more trawling revealed this chart for a Nikon F3 film camera:

https://www.mir.com.my/rb/photograph...l/EV_Chart.pdf

Scary stuff. For example it appears that if you load 400 ASA film and set f/11 - you run out of shadow metering at 9EV which is like twilight.

Just when I thought I knew everything about metering ...

FWIW, it turns out that the complicated-looking chart is just a set of graph lines for the EV exposure formula:

EV=log2(N^2/t) where N^2 = f-number squared and t = shutter-time.

So not so hard as I thought.

[Manfred M]

One thing that photographers do not seem to recognize is what a "proper exposure" really is.

When I teach photography basics I get people to photograph the same scene at different focal lengths (my setup uses a tripod) and I will ask why the (matrix) exposure changes as one zooms in.

I tend to use a lens with a zoom factor that is greater than 10:1 (something like my 28 - 300mm lens). Some people get it while others are a bit shocked that the actual exposure does change. I generally do this outdoors and ensure that the wide angle shot includes a lot of sky and the tight shot is of a fairly dark area.

When it comes to spot metering, people don't necessarily understand that the "spot" is going to give an 18% gray (or whatever the camera manufacturer happens to have used) versus the "correct" overall exposure. I have a 2 degree spot metering head for my Sekonic L-358 meter and let people go wild with it.

I spend a lot of time acquainting people with the histogram displays on their camera.

I don't get into heavy theory because most people don't care. They just want to get a good image,

[xpatUSA]

One thing that photographers do not seem to recognize is what a "proper exposure" really is.

When I teach photography basics I get people to photograph the same scene at different focal lengths (my setup uses a tripod) and I will ask why the (matrix) exposure changes as one zooms in.

In support of your post, there's a really good visual explanation of the three major forms of metering on document page 40 here:

http://www.alstonhand.com/ueditor/js...1349373607.pdf
.

[DanK]

I just looked up the specs for my Canon 5D Mark IV. On the German site, it says EV 0-20 for evaluative metering and doesn't mention the others. On the US site, it says EV 0-20 as a general statement, with no mention of it's applying to one or more metering modes.

In addition to the good advice offered so far and after looking up Metering Range with reference to scene brights and darks, I found something against spot metering.



https://www.imaging-resource.com/PRO...00/D7000A5.HTM

Meaning that Spot is not so hot when pointed at shadows. I never knew that. Grump.

Then I looked the metering range for my Sigma SD15 DSLR (separate metering sensor under the mirror). It is 1 to 20 EV but ominously specified at 100 ISO, 50mm, f/1.4 - almost certainly meaning less at other settings but no examples given.

So, more trawling revealed this chart for a Nikon F3 film camera:

https://www.mir.com.my/rb/photograph...l/EV_Chart.pdf

Scary stuff. For example it appears that if you load 400 ASA film and set f/11 - you run out of shadow metering at 9EV which is like twilight.